1. Field of the Invention
The present invention relates to an image processing technique and, more particularly, to a range setting technique of a reading region.
2. Description of the Related Art
Conventionally, image processing apparatuses which read originals placed on original plates of a scanner, copying machine, and the like as digital images are available. In an original detection method using an image processing apparatus, an original region is detected from the density distribution of image information after the entire surface of the original plate is read. According to this image processing apparatus, the position of an original placed on the original plate is automatically detected. For this reason, the user can specify a region to be read and can acquire a desired image without any operations for designating the region to be read or placing an original at a predetermined position on the original plate.
One may often want to read a specific information region alone from an original which includes specific text information (e.g., a personal name, telephone number, and e-mail address) like a business card. Sometimes, the contents of an original having a predetermined format such as a ledger sheet may include description errors. In this case, one may require an original form without any entries and may want to designate a part (format part in this case) of the original as a region to be read. In such a case, as a method of designating a specific region in a detected original, Japanese Patent Laid-Open No. 2003-32488 (reference 1) discloses a technique for embedding, in an original, position information from one end of the original, and specifying a region based on the position information.
Furthermore, when an original digital file of a paper original read by a scanner is available, one may want to save only the modified part of the read paper original. Japanese Patent Laid-Open Nos. 2002-278984 (reference 2) and 2004-080601 (reference 3) disclose techniques for detecting whether or not a paper original is modified by calculating the difference between image data generated based on an original digital file and that read by the scanner.
However, with the technique described in reference 1 above, when a change in density distribution between the original plate and an image printed on an original is small, the detection precision tends to drop. For example, when an arbitrary image is printed on the entire surface of an original like a photo or leaflet, since there is a density difference between the original plate and the original, detection of the boundary between the original plate and the original easily succeeds. On the other hand, in case of an original printed on a paper sheet which has a hue close to that of the original plate, detection of the boundary between the original plate and the original often fails. When detection of the boundary between the original plate and the original fails, the related art which holds a specific region in an original as position information from one end of the original cannot determine the specific region.
With the technique described in reference 2 or 3 above, an original digital file must be converted into image data to calculate the difference between the original digital file and read image data, resulting in a heavy processing load and long processing time.